Which AI tools does the MCP server work with?

With all common MCP clients: Claude, ChatGPT, Claude Code, Codex, Cursor, Claude Desktop, Windsurf, Cline, Continue, VS Code and Zed. If your tool supports the Model Context Protocol, you can connect Famulor.

How secure is the connection?

Sign-in runs over OAuth, so you never share passwords or API keys in plain text. You pick one workspace and approve its tool groups: read-only, read and write, or a custom selection. Revoke any connection under Settings → API & MCP. For server-to-server integrations, use workspace-scoped API keys with their own permissions and expiry.

How do I set up the MCP server?

To review and analyze, install the official Famulor plugin in ChatGPT or Codex. For full control, add a custom connector with the URL from Settings → API & MCP, sign in via OAuth and approve the tool groups. Guide: https://docs.famulor.io/mcp/overview.

Do I need programming skills?

No. You control everything in plain language. Developers get a bonus, they can prototype assistants directly in Cursor or Claude and bulk-create campaigns from CSV, but it is not required.

What does the official Famulor plugin in ChatGPT and Codex do?

The plugin shows assistant configurations, saved versions, available models, languages and voices, plus conversation history and transcripts. Review setups and ask for suggested improvements without leaving ChatGPT or Codex. To change assistants, place calls or run campaigns, use a custom MCP connection with the URL from Settings → API & MCP.

Can I ask Milian from my AI client?

Yes. Select the optional Milian tool group when you connect and use ask_milian, or call POST /api/v1/milian/ask on the REST API. Milian answers questions, analyses the workspace data your connection permits and recommends next steps without changing resources. Each question uses additional workspace credits at the same rates as the dashboard, and your client asks for your explicit approval first.

Famulor MCP Server: Run Voice Agents

Name: Famulor
Brand: Famulor
Price: 0.11 EUR
Availability: InStock

READY IN 3 STEPS

Pick your path: MCP, CLI or Skill.

Open Claude connectors

Launch the app or open claude.ai and go to Customize → Connectors.

Customize → Connectors

Add the Famulor custom connector

Name it Famulor and paste the URL:

https://app.famulor.io/mcp?profile=assistant-history

Connect and sign in

Approve read access, then ask: “Summarize my latest calls.”

Using Claude Code, Codex, Gemini CLI, OpenClaw or Hermes?

MCP CLIENTS

Connect Famulor to your AI tool.

Add the Famulor MCP server in your favourite client, via one-click install or our setup guide.

Serverhttps://app.famulor.io/mcp

Claude Code
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Claude
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Cursor
- Voice agents, calls & campaigns
- One-click install
Add MCP to Cursor
VS Code
- Voice agents, calls & campaigns
- One-click install
Add MCP to VS Code
ChatGPT
- Voice agents, calls & campaigns
- One-click install
Open in ChatGPT
Codex
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Windsurf
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Kiro
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Replit
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Warp
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Firebender
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Factory
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Lovable
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Augment
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Google Gemini CLI
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Atlassian Studio
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Android Studio
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Amazon Q
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
OpenHands
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Zed
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
ServiceNow Build Agent
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
OpenClaw
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Hermes
- Voice agents, calls & campaigns
- Remote MCP over HTTP
View setup guide
Famulor Skill
- Voice agents, calls & campaigns
- For Claude Code, Cursor, OpenClaw & more
npx skills add bekservice/Famulor-Skill
Install Famulor Skill

Your tool missing? Any MCP-capable client works. Famulor Skill on GitHub

Highlights

What the agent does here

The parts of this feature teams use first on a live call.

Build voice agents by prompt

Say what you need. MCP creates the assistant, sets prompt and knowledge, no dashboard clicks.

Place and analyze calls

"Call this number" or "show today's missed-goal calls": right in your AI client.

Run campaigns from chat

Create a campaign, start it, add leads from CSV, by voice, not click-config.

Results

Numbers that matter

MCP server for your whole account

controllable modules & areas

15+

secure sign-in

OAuth

natural language

100%

Customer story

“We now build and launch our voice agents entirely from Claude. What took about an hour in the dashboard is one sentence in chat: we shipped 14 outbound campaigns in a single week and tripled our setup speed.”

Marcus Whitfield

VP Revenue Operations, Cedarline Solutions (Austin, TX)

Features

What ships with this feature

Each part of this feature, ready to run together from day one.

Build voice agents by prompt

Say what you need. MCP creates the assistant, sets prompt and knowledge, no dashboard clicks.

Place and analyze calls

"Call this number" or "show today's missed-goal calls": right in your AI client.

Run campaigns from chat

Create a campaign, start it, add leads from CSV, by voice, not click-config.

Maintain leads & knowledge

Create leads, update fields, scrape URLs and link knowledge: one command.

Telephony, SIP & multi-channel

Numbers, SIP, WhatsApp and SMS from chat, or turn AI off for a human handoff.

At home in every MCP client

Claude, Cursor, Codex and more, plus the official Famulor plugin in ChatGPT and Codex.

How it works

Live in a few steps

No lengthy setup: most teams are live the same week.

01
Connect your AI client
Install the official Famulor plugin in ChatGPT or Codex to review assistants and history. For full control, add the URL from Settings → API & MCP in Claude, ChatGPT, Cursor or another MCP client and sign in via OAuth.
02
Instruct in natural language
Just describe what you need: "Build me an inbound assistant", "Start the campaign" or "Show me today's calls". The MCP server turns your command into real actions.
03
Results live in your account
Assistants, calls, campaigns and leads are created and updated in your Famulor account instantly. Pull reporting like credit balance and call stats the same way, by prompt.

Live Demo

Sounds unreal? Try it now with a free test call

Curious how our AI agents work?

Get a hands-on experience by trying a free demo call. Fill in your details, and our AI representative will call you instantly.

Agent is trained to discuss Famulor services and book appointments.

Demo AI agent

Famulor representative

🇺🇸English

Works with

Questions before you go live

Practically your entire account: create and configure voice agents, place calls, manage campaigns and leads, maintain knowledge bases, use WhatsApp/SMS/chat, provision telephony and SIP trunks, and set up mid-call integrations and webhooks: all in natural language.

Dive into Famulor MCP

Get the Famulor MCP setup guide

Learn how to activate the Famulor MCP server remotely or in-app, add clients, and write powerful prompts.

Read the guide

Connect your setup with Dev Mode

The Famulor MCP works even better in your dev workflow. Connect your code with the Famulor Skill for sharper results.

Learn more

Introducing: the Famulor MCP server

Bring Famulor context straight into your dev workflow, so LLMs generate real, executable actions and code.

Read the article

Get started

Connect Famulor to your AI client today.

Run your entire account in plain language from Claude, ChatGPT, Cursor and more: secured via OAuth, without a single click in the dashboard.

No credit card required · GDPR compliant · first agent live in under a week

READY IN 3 STEPS